AWS Bedrock

Connect your AI agents to foundation models available through AWS Bedrock, including Amazon Nova, Anthropic Claude, and other providers.

Overview

The AWS Bedrock connector lets you access foundation models through Amazon Bedrock. This is useful when your infrastructure is on AWS and you want to use Bedrock's managed model access, or when you need access to models available exclusively through Bedrock such as Amazon Nova.

The connector supports:

• Chat completions using the Bedrock Converse API
• Function calling (tool use)

Setting up the connector

To add an AWS Bedrock connector, complete the following steps:

1. Navigate to the Squid Console and select your application.
2. Click the Connectors tab.
3. Click Available Connectors and find the AWS Bedrock connector. Then click Add Connector.
4. Provide the following details:

• Connector ID: A unique ID of your choice (e.g., my-bedrock). This is the integrationId you will reference in code.
• Region: The AWS region where your Bedrock service is available (e.g., us-east-1, us-west-2). Ensure the models you want to use are enabled in this region.
• API Key: Your AWS Bedrock API key.
• Models: A JSON array defining the models available through this connector. Each model requires the following fields:

Field	Type	Description
modelName	string	The Bedrock model identifier (e.g., us.amazon.nova-2-lite-v1:0)
displayName	string	A human-readable name for the model
maxOutputTokens	number	Maximum number of tokens the model can generate in a response
contextWindowTokens	number	Total context window size in tokens

Example:

[
  {
    "modelName": "us.amazon.nova-2-lite-v1:0",
    "displayName": "Amazon Nova 2 Lite",
    "maxOutputTokens": 5120,
    "contextWindowTokens": 300000
  }
]

5. Click Add Connector.

Using the connector

Once configured, use the connector with an AI agent by specifying the connector ID and model name:

Client code

await squid.ai().agent('my-agent').updateModel({
  integrationId: 'my-bedrock',
  model: 'us.amazon.nova-2-lite-v1:0',
});

You can also override the model on a per-request basis:

Client code

const response = await squid
  .ai()
  .agent('my-agent')
  .ask('Summarize our Q4 report', {
    model: {
      integrationId: 'my-bedrock',
      model: 'us.amazon.nova-2-lite-v1:0',
    },
  });

Embeddings

If your Squid application is running in one of Squid's AWS regions, Amazon Titan Text Embeddings v2 (titan-embed-text-v2) will appear as an available embedding model by default when creating a knowledge base. This is separate from the Bedrock connector itself and does not require any additional configuration.